Sturge–Weber syndrome & port-wine stain: Genomic analysis reveals somatic activating mutation in GNAQ as genetic basis for neurocutaneous disorder
Matt Shirley
22 May 2013
Pevsner lab projects
- Identify regions and magnitude of structural variation in LCLs:
- Determine the genetic basis for Sturge-Weber syndrome
Outline
- What is the Sturge-Weber syndrome?
- Sequencing experiments
- Molecular biology of GNAQ mutations
- GNAQ in SWS etiology: regulated or regulator?
- Conclusions
Pathology of the Sturge-Weber syndrome
- The Sturge-Weber syndrome (SWS):
- Port-wine stain (PWS) affecting the face
- Abnormal capillary blood vessels in the brain
- Results in seizures, stroke, and intellectual disability
- Patients may develop abnormal capillaries in the eye (glaucoma)
How common is Sturge-Weber syndrome?
- Approximately 1 in 20-50,000 live births
- Probability of developing SWS is 6-26% for children with a facial port-wine stain (PWS)
- PWS occurrence is approximately 3 in 1000 in the general population
Representative images of subjects from this study
Outline
- What is the Sturge-Weber syndrome?
- Sequencing experiments
- Molecular biology of GNAQ mutations
- GNAQ in SWS etiology: regulated or regulator?
- Conclusions
Exploratory sequencing from paired whole genomes
Paired sample collection + >30X whole genome sequencing:
- Identify all mutations in a normal sample
- Identify all mutations in the affected sample
- Combine the two signals, accounting for some level of background noise in the data
For somatic mutation detection from paired samples we use Strelka.
How do we identify the correct somatic mutation?
Using Strelka, we identified somatic SNVs in each affected tissue
| 1 |
325 |
PWS |
| 2 |
543 |
Brain |
| 3 |
427 |
PWS |
- Strategies for variant prioritization:
- Functional annotation of the combined SNV call set using VAAST (Variant Annotation and Search Tool)
- Expansion of the call set by examining variant allele frequencies at lower thresholds
Functional annotation of variants using VAAST
Variants in → disease alleles out
1294 variants → 1 candidate variant: GNAQ R183Q
Expanding our list of shared somatic variants
Blue box: variants having a mutant allele in affected sample and not in normal tissue samples
M = sum of log transformed allele frequencies
A = difference of log transformed allele frequencies
Set intersection of low-confidence somatic variants still identifies GNAQ
GNAQ is still the only shared coding somatic variant
GNAQ is a G-protein alpha subunit
Image from EuroEPINOMICS
GNAQ somatic mutations are associated with uveal melanomas and melanocytic lesions
- This R183Q mutation as well as Q209L are associated with uveal melanomas when occuring in melanocytes
- These same mutations can also cause non-cancerous melanocytic lesions (blue nevus and nevus of Ota)
- There are rare occurrences of SWS and melanocytic lesions (Phakomatosis pigmentovascularis)
Targeted deep amplicon sequencing design: primers
Custom PCR amplicon sequencing with Illumina MiSeq
- Designed a custom amplicon sequencing strategy
- Illumina's amplicon kits cost ~10X more
- Incorporated error-correcting barcodes described by Leonid Bystrykh.
- Wrote a program to encode, decode, and error-correct these barcodes
- Error correction allows us to acheive un-ambiguous multiplexing of our samples during one sequencing run
- In practice this also allowed us to acheive 4-9% greater read depth than without error correction
Error correction barcodes
in fixed checksum
CATAACT CAAAACT A > T at pos 3
AGAGAGA AGAGAGA ok
TCACAGC TCACAGC ok
GAACAGG GAAAAGG A > C at pos 4
CTATAGT CTATAGT ok
TTTAAAN NNNNNNN bad
NNNNNNN NNNNNNN bad
Targeted deep amplicon sequencing design: read depth
Based on this simulation, we targeted > 10,000 and achieved 2446 to 93,008 (median, 12,947) read depth.
Results: GNAQ somatic mutation is found in PWS and SWS brain
| 9 |
PWS |
Yes |
100% |
Amplicon seq |
| 7 |
Skin |
Yes |
14% |
Amplicon seq |
| 13 |
PWS |
No |
92% |
Amplicon seq, Primer extension |
| 18 |
Brain |
Yes |
88% |
Amplicon seq |
| 6 |
Brain |
No |
0% |
Amplicon seq |
| 4 |
Brain |
CCM |
0% |
Primer extension |
| 669 |
Blood/LCL |
N/A |
0.7% |
Exome seq |
CCM = cerebral cavernous malformation
Amplicon seq = 12,947X median read depth
Exome seq = 271X median read depth
Primer extension = SNaPshot allele-specific primer extension
Outline
- What is the Sturge-Weber syndrome?
- Sequencing experiments
- Molecular biology of GNAQ mutations
- GNAQ in SWS etiology: regulated or regulator?
- Conclusions
Effect of R183Q GNAQ mutation on MAPK signaling pathways
- Trans-expression of mutant GNAQ activates MAPK signaling pathways
- R183Q mutation is a less potent activator than Q209L
- R183Q only activates ERK and JNK
Outline
- What is the Sturge-Weber syndrome?
- Sequencing experiments
- Molecular biology of GNAQ mutations
- GNAQ in SWS etiology: regulated or regulator?
- Conclusions
GNAQ regulation by Regulator of G-protein signaling (RGS)
RGS4 serves as a GTPase-activating protein for GNAQ
- RGS4 cannot regulate Q209L GNAQ (homologous to mouse Q204L), but regains the ability to regulate R183Q (homologous to R178C)
Berman 1996 Cell
GNAQ interaction with Trio (GEF) activates MAPK pathway independent of PLC-beta
- Knockdown of Trio does not affect ERK activity
- Does GNAQ R183Q prevent this interaction?
Vaqué 2012 Molecular Cell
Conclusions / Implications
- Identified the genetic basis for Sturge-Weber syndrome and non-syndromic port-wine stains
- Demonstrated that GNAQ R183Q somatic activating mutation has a modest and differential effect on MAPK signal transduction
- These findings may also apply to Klippel-Trenaunay syndrome (a more extensive bodily PWS)
- GNAQ somatic activating mutations in different cell types give rise to different phenotypes
- R183Q mutation in a melanocyte gives rise to uveal melanoma or melanocytic lesions
- R183Q mutation in vascular endothelial cell (yet to be identified) gives rise to PWS or SWS
- Targeted pharmaceutical therapies for PWS and SWS may now be possible
Questions?
Acknowledgements
Lab members
Dr. Jonathan PevsnerDr. Joseph Baugher
Dr. Eric Stevens
Larry Frelin
Don Freed
Amanda and Genevieve
Thesis committee members
Dr. Garry Cutting
Dr. Dani Fallin
Dr. Sarah Wheelan
Dr. Ingo Ruczinski
Collaborators
Dr. Anne Comi (Kennedy Krieger Institute)
Dr. Douglas Marchuk (Duke University)
Dr. Bernard Cohen (Kennedy Krieger Institute)
Dr. Paula North (Medical College of Wisconsin)
Dr. Hao Tang (Duke University)
Carol Gallione (Duke University)
Patients and families
BCMB program at Hopkins
Dr. Carolyn Machamer
Margie Policastri
Sharon Root
Dr. Leslie Brown
Dr. Arhonda Gogos
Our growing family
